Familiarity modeling

• US 9,417,069 B2
• Filed: 07/25/2013
• Issued: 08/16/2016
• Est. Priority Date: 07/25/2013
• Status: Expired due to Fees

• Current Assignee: Honda Motor Co., Ltd. (Honda Motor Company)
• Original Assignee: Honda Motor Co., Ltd. (Honda Motor Company)
• Inventors: Gupta, Rakesh, Karpov, Igor V., Raux, Antoine, Ramachandran, Deepak
• Primary Examiner(s): Shah, Kamini S
• Assistant Examiner(s): Louis, Andre Pierre
• Application Number:
  US13/951,015
• Publication Number:
  US 20150032424A1
• Time in Patent Office: 1,118 Days
• Field of Search: 703/2
• US Class Current: 1/1
• CPC Class Codes: G01C 21/3484 Personalized, e.g. from learned user behaviour or user-defined profiles
• CPC Class Codes: G01C 21/3641 Personalized guidance, e.g. limited guidance on previously travelled routes
• CPC Class Codes: G06N 20/00 Machine learning

Abstract

One or more embodiments of techniques or systems for modeling familiarity for a traveler are provided herein. Familiarity evidence can be received, indicative of how familiar a traveler is with an area or road segment, and based on a number of visits the traveler has made to that area. The familiarity evidence can be used to generate one or more familiarity models indicative of a predicted familiarity of locations around the area. Familiarity models can be based on kernels, graph distances, Markov random fields (MRFs), etc. When route directions are generated from an origin location to a destination location, one or more of the directions can be provided based on one or more of the familiarity models. For example, if a familiarity model indicates that a traveler is familiar with a route, driving directions of the route can be adapted to be more succinct.

Claims

1. A method for modeling familiarity for a traveler, comprising:

receiving familiarity evidence associated with one or more road segments of a map and one or more intersections of the map, wherein one or more of the intersections are connected via one or more of the road segments, wherein the familiarity evidence comprises one or more familiarity scores for one or more of the road segments or one or more of the intersections;

generating a road network graph (RNG) comprising one or more RNG nodes corresponding to one or more of the intersections of the map and one or more RNG edges corresponding to one or more of the road segments of the map;

generating a Markov random field (MRF) graph based on the RNG by connecting the RNG edges that are incident at a same intersection to form one or more nodes of the MRF graph and connecting at least a portion of the nodes of the MRF graph when corresponding edges of the RNG share a common node; and

generating one or more familiarity models for the map based on the familiarity evidence, wherein one or more of the familiarity models are based on the MRF graph; and

in response to detecting a difference between familiarity scores assigned to two nodes of a clique of the MRF graph exceeding a selected threshold in one of the familiarity models based on the MRF graph, adjusting the familiarity scores assigned to the two nodes to reduce the difference, wherein the adjusting comprises setting a fuel gauge or energy meter of a vehicle driven by the traveler to read lower than the actual reading of the vehicle fuel or energy level when the traveler is driving the vehicle in an unfamiliar area,

wherein the receiving and the generating are implemented via a processing unit.

2. The method of claim 1, wherein one or more of the familiarity models are further based on one or more kernels or one or more graph distances, and one or more of the familiarity models comprise one or more predicted familiarity scores for one or more of the road segments or one or more of the intersections.

3. The method of claim 1, wherein the familiarity evidence is indicative of how frequently the traveler travels along one or more of the road segments or one or more of the intersections.

4. The method of claim 2, wherein one of more of the graph distances is measured by nodes, edges, travel time, or major intersections.

5. A system for modeling familiarity for a traveler, comprising:

a collection component configured to receive familiarity evidence associated with one or more road segments of a map and one or more intersections of the map, wherein one or more of the intersections are connected via one or more of the road segments, wherein the familiarity evidence comprises one or more familiarity scores for one or more of the road segments or one or more of the intersections; and

a familiarity model component configured to;

generate a road network graph (RNG) comprising one or more RNG nodes corresponding to one or more of the intersections of the map and one or more RNG edges corresponding to one or more of the road segments of the map;

generate a Markov random field (MRF) graph based on the RNG by connecting the RNG edges that are incident at a same intersection to form one or more nodes of the MRF graph and connecting at least a portion of the nodes of the MRF graph when corresponding edges of the RNG share a common node;

generate one or more familiarity models for the map based on the familiarity evidence, wherein one or more of the familiarity models are based on the MRF graph, wherein one or more of the familiarity models comprise one or more predicted familiarity scores for one or more of the road segments or one or more of the intersections;

in response to detecting a difference between familiarity scores assigned to two nodes of a clique of the MRF graph exceeding a selected threshold in one of the familiarity models based on the MRF graph, adjust the familiarity scores assigned to the two nodes to reduce the difference and adjust a fuel gauge or energy meter of a vehicle driven by the traveler to read lower than the actual reading of the vehicle fuel or energy level when the traveler is driving the vehicle in an unfamiliar area,

wherein the collection component or familiarity model component is implemented via a processing unit.

6. The system of claim 5, wherein one or more of the familiarity models are further based on one or more kernels or one or more graph distances, and the familiarity model component is configured to generate one or more potential functions over one or more of the MRF nodes of a clique or one or more of the MRF edges of the clique.

7. The system of claim 6, wherein one or more of the potential functions is associated with one or more aspects of the RNG.

8. The system of claim 5, further comprising:

a telematics component configured to capture global positioning system (GPS) data; and

a score component configured to assign one or more of the familiarity scores to one or more of the road segments or one or more of the intersections based on the GPS data.

9. The system of claim 8, wherein the telematics component is configured to:

receive an origin location;

receive a destination location; and

generate a route from the origin location to the destination location, wherein the route comprises one or more route segments corresponding to one or more of the road segments of the map.

10. The system of claim 9, further comprising a cognition component configured to provide one or more directions from the origin location to the destination location based on the route and one or more of the familiarity models.

11. A non-transitory computer-readable storage medium comprising computer-executable instructions, which when executed via a processing unit on a computer performs acts, comprising:

receiving familiarity evidence associated with one or more road segments of a map and one or more intersections of the map, wherein one or more of the intersections are connected via one or more of the road segments, wherein the familiarity evidence comprises one or more familiarity scores for one or more of the road segments or one or more of the intersections;

generating a road network graph (RNG) comprising one or more RNG nodes corresponding to one or more of the intersections of the map and one or more RNG edges corresponding to one or more of the road segments of the map;

generating a Markov random field (MRF) graph based on the RNG by connecting the RNG edges that are incident at a same intersection to form one or more nodes of the MRF graph and connecting at least a portion of the nodes of the MRF graph when corresponding edges of the RNG share a common node;

generating one or more familiarity models for the map based on the familiarity evidence, wherein one or more of the familiarity models are based on the MRF graph and comprise one or more predicted familiarity scores for one or more of the road segments or one or more of the intersections;

in response to detecting a difference between familiarity scores assigned to two nodes of a clique of the MRF graph exceeding a selected threshold in one of the familiarity models based on the MRF graph, adjusting the familiarity scores assigned to the two nodes to reduce the difference, wherein the adjusting comprises setting a fuel gauge or energy meter of a vehicle driven by the traveler to read lower than the actual reading of the vehicle fuel or energy level when the traveler is driving the vehicle in an unfamiliar area;

receiving a route from an origin location to a destination location, the route comprising one or more route segments corresponding to one or more of the road segments of the map; and

providing one or more directions for one or more of the route segments based on the familiarity models which are based on the MRF graph.

12. The non-transitory computer-readable storage medium of claim 11, wherein one or more of the familiarity models are further based on one or more kernels or one or more graph distances, and the method further comprises instructions for:

receiving feedback associated with one or more of the directions; and

adjusting the map kernel familiarity model, the graph distance familiarity model, or the MRF familiarity model based on the feedback.

13. The non-transitory computer-readable storage medium of claim 12, further comprising selecting the route from the origin location to the destination location based on the map kernel familiarity model, the graph distance familiarity model, or the MRF familiarity model.

14. The non-transitory computer-readable storage medium of claim 11, wherein the familiarity evidence comprises global positioning system (GPS) data.

15. The non-transitory computer-readable storage medium of claim 11, wherein the familiarity evidence is indicative of how frequently a traveler travels along one or more of the road segments or one or more of the intersections.

16. The non-transitory computer-readable storage medium of claim 11, further comprising generating one or more potential functions for one or more cliques associated with the MRF familiarity model.